The federal loan repayment program for food-animal veterinarians offsets debt rather than revenue — United States farms with cattle and calves fell 38 percent from 1997 to 2022 while the herd fell only 12 percent

What is happening?

In 2026 the United States Department of Agriculture designated 245 veterinary shortage areas across 47 states, a figure published on 2026-02-02. The preceding set, for FY2025, was 243 areas across 46 states and was described as a record. The set before that, in 2024, was 240 across 47 states. Growth runs at roughly two areas a year on a base of 240, so the striking quantity here is the level rather than the trend.

The federal instrument built to answer that list is the Veterinary Medicine Loan Repayment Program, authorized by 7 U.S.C. 3151a and governed by 7 CFR part 3431. In FY2024 it obligated USD 11.1 million to 114 veterinarians — 80 of them new placements and 34 renewals of service agreements already running. Cumulative awards since 2010 total 997, which averages about 62 a year including renewals; that average is derived arithmetic and is printed by no source.

The designation count is not a headcount. Form NIFA 2009-0001, the nomination form that creates a designation, records must-serve and may-serve locations, priority, practice type, species, narrative need, past recruitment efforts and the consequences of leaving the post unfilled. It never asks how many veterinarians the area needs, and it records no count of how many prior years the same area was nominated. So 245 counts situations rather than vacancies, and the list cannot be summed into a national gap.

Whose problem is this?

The body that designates the shortage areas is the same body that funds the awards, and the two numbers it produces have never been tied to each other. The fix that would enlarge the awards without new money sits with a different set of committees entirely.

Where does this problem end?

The boundary matters here because the instrument already exists and has run for sixteen award cycles. What is missing is not a program but a size, and nothing in statute or budget process ties the size of the program to the length of the list.

What is the state now, and what should it be?

Now

Needs a new measurementthe target state: no source opened here sets a target for the number of designations, for the share of designated situations that receive an award, or for the food-animal share of the veterinary workforce. The Rural Veterinary Action Plan of 2025-08-28 lists five actions and no number to reach, and the appropriation is a dollar figure with no denominator attached to it.

How big is it?

Between about 1.28 million and about 3.33 million producers, as of the 2022 Census of Agriculture. That range is the national client base of food-animal veterinary practice. It is not a count of people inside a designated shortage area, and it must not be read as one.

The low bound is producers on the 732,123 United States farms that had cattle and calves in 2022, multiplied by a derived ratio of about 1.75 producers per farm. That ratio comes from dividing about 3.33 million producers, itself derived from the 1.2 million female producers reported as 36 percent of all producers, by 1,900,487 farms. The product is about 1,281,000 people. The high bound is all 3.33 million United States farm producers, the outer envelope once hog-only, poultry-only and small-ruminant operations and their co-operators are folded in. Both the ratio and the product are arithmetic performed here, not figures published by any agency.

The in-area count is not derivable, and the missing input is specific. USDA publishes the designations as a map application rather than as a county roster, and 7 CFR 3431.7 places them on an agency web site rather than in the Federal Register, so there is no dated roster of which counties were designated in any given year. Designations sit in 47 of 50 states, which makes the national base the right universe to bound. But the share of that base inside a designated area cannot be computed from any published source, and no proxy share is defensible, so none is offered here.

The range also fails to count several groups entirely. The general stake in animal-disease surveillance and in the lead time for detecting a foreign animal disease is asserted by USDA and quantified nowhere. The animals are not counted, and 86.2 million cattle stood in national inventory on 2026-01-01. Meat and poultry inspection and export certification capacity sits behind public-practice designations whose clients are regulatory functions rather than farms. Rural residents who are not producers but live in a county whose only large-animal practice closed are outside every count used here. And the 3,424 veterinarians in food-animal clinical practice are excluded on purpose, because they are the missing service rather than the affected population. All of that sits on top of a bound that already cannot be narrowed to the designated areas themselves.

Under what conditions does it arise?

1. The list is capped before it is compiled. 7 CFR 3431.5(d) lets the solicitation set a maximum number of nominations per State Animal Health Official, and the nomination form directs nominators to the allocation table for their own limit. A peer panel then screens what is submitted, under 7 CFR 3431.6(a), before the Secretary designates. The published count is therefore a function of the per-state cap, of nomination behaviour and of panel scoring, and not of any survey of need. The allocation table itself could not be opened this round, so the national ceiling is not established here.

2. The unit has no denominator. One designation is one situation, whatever the size of its service area, because the form records no headcount. Designations also come in three types, and Type III covers public practice — government and academic posts in food safety, public health and epidemiology rather than farm calls. The split among the three types is published nowhere, so the private food-animal subset of 245 is unknown.

3. The instruments are sized by appropriation, not by the list. About USD 18 million funds 80 to 120 awards a year, a third of them renewals. Nothing in statute, rule or budget process ties the number of awards to the number of designations, so the two are free to drift apart indefinitely. The second instrument is smaller still, at 14 practice grants in the round announced on 2025-12-17.

4. A fixed share of every repayment dollar returns to the Treasury. 7 CFR 3431.13(d) sets the tax payment at 39 percent of the loan repayment, which is why an award of USD 40,000 a year carries a further USD 15,600 a year. In FY2024 that came to USD 3.1 million against USD 11.1 million obligated. The 39 percent is a share of loan repayments and not a share of the appropriation.

5. The pipeline runs fifteen months before day one. Nominations for the FY2026 cycle opened 2025-09-22 and closed 2025-11-07, shortage areas were posted 2026-01-05, applications ran from 2026-01-13 to 2026-03-05, awards are offered 2026-09-30, and the service agreement begins 2027-01-01. A situation nominated in autumn produces a working veterinarian more than a year later, at the start of a three-year obligation.

What has been tried?

Two directions have been tried at once, and both work on the same side of the ledger. One raises the award, the other adds a grant program, and neither changes what a rural practice can collect once the three-year obligation ends. A third direction, removing the tax that consumes 39 percent of each repayment dollar, has been introduced repeatedly and never enacted.

What was found?

Why is it still unsolved?

Cost structure — the revenue a rural food-animal practice can collect has fallen faster than the animals it serves, and every federal instrument attached to the problem offsets debt rather than revenue.

The first movement is that there is no absolute deficit of veterinarians. 86,251 were in clinical practice in 2024, and over the decade to 2023 the companion-animal segment grew 22 percent while food and mixed animal practice fell 15 percent. The profession did not shrink. One segment of it emptied, because that segment cannot bid for the graduates it needs against the segment next door.

The second movement is on the client side, and it is the decisive pair of numbers. Between 1997 and 2022 the number of United States farms with cattle and calves fell 38 percent, from 1,188,659 to 732,123, while the cattle and calves inventory fell only 12 percent. Subtracting those two counts leaves 456,536 fewer paying operations behind substantially the same herd, spread across the same geography. That is not a statement about where veterinarians live. It is a statement about how much revenue a practice can collect per mile driven, and it moved against food-animal practice for twenty-five years before any of the current instruments existed.

The third movement is what the instrument can and cannot reach. Three years of USD 40,000 a year ends, and what remains is about USD 100,000 in mixed-animal associate pay in a service area with 38 percent fewer paying operations than a generation ago, against USD 133,000 for companion-animal work in a metropolitan area, carried against over USD 212,000 of educational debt. The subsidy addresses the debt and cannot touch the revenue, and it does not try to. Meanwhile 39 percent of each repayment dollar is paid straight back into a federal tax account by regulation, and the exclusion that physicians already have under 42 U.S.C. 254l-1 has been introduced in at least four Congresses since 2015 without being enacted. That fix belongs to the tax-writing committees, while the program belongs to the agriculture committees, so the instrument and its largest single leak are owned by different rooms.

A close runner-up describes the same file accurately at the level of observation. The service is unavailable in one segment and one geography while the profession expands elsewhere, which is a disparity. It is not taken as the primary reading because it names the distribution rather than the cause, and the cause is that the segment cannot compete on price for the people it needs.

What observation would mean it is solved?

Candidates — (a) the number of designated shortage situations falls for several consecutive years (b) the share of designated situations that receive an award in the same cycle rises toward one (c) the food-animal share of the veterinary workforce stops falling and turns up.

(a) alone is the weakest. The count is bounded above by a per-state nomination cap and produced by nomination behaviour, so a fall could mean areas were filled, or that states nominated fewer, or that the solicitation ran on the two or three year cycle that 7 CFR 3431.5(c) permits. A record and a ceiling look identical from outside when the ceiling is not published.

(b) alone would first have to exist. No fill rate is published today, so adopting this observation would require USDA to start publishing one, and the annual report that would carry it currently stops at FY2024. An award is also a three-year obligation rather than a permanent practice, and nothing federal is published at year four. Kansas publishes 94 percent retention at four years for its own program, which is the shape of the missing number.

(c) alone counts headcount and not service. The 3,424 food-animal veterinarians of 2024 does not say where they are, which species they serve, or how many hours of the week go to food supply work. The two published workforce categories do not reconcile with each other, so a rise in one could be a definitional artefact rather than a change on the ground.

The three have to be read together, and all three have to be read against the client base, which fell 38 percent in operations while the herd fell 12 percent.

What is it connected to?

Fills with researchrural hospital and rural pharmacy access, veterinary school capacity and tuition, foreign animal disease preparedness and border inspection staffing, meat and poultry inspection capacity, and the design of the same instrument in human medicine under 42 U.S.C. 254l-1. Relation type and evidence grade were not confirmed in this round.

What these sources do not say

• No need-quantum exists anywhere. Neither 7 CFR part 3431 nor Form NIFA 2009-0001 asks how many veterinarians a shortage situation requires. The list counts situations. Not one source that repeats the figure of 245 shortage areas notes that the number is not a headcount.
• The ceiling is never mentioned beside the record. Each state is capped by a NIFA allocation table. No source calling 243 or 245 a record mentions that a cap exists, and the allocation total was not reachable from any route tried this round.
• The type split is never published. Type III designations are public practice, covering regulatory, food-safety, epidemiology and academic posts rather than farm calls. No source that quotes the headline count publishes how many of the 245 are Type I, Type II or Type III, so everyone describing all 245 as food-animal areas overstates the private-practice subset by an unknown amount.
• No fill rate exists in public. Awards are counted every year, at 114 in FY2024. How many designated situations received an award is published nowhere. The one number that would say whether the program works against its own list has never been stated.
• No re-designation series exists, and the list is never archived. Nothing records how many consecutive years the same area has been designated, and the form stores past recruitment efforts only as prose. Because 7 CFR 3431.7 puts designations on a web site rather than in the Federal Register, there is also no archival record of what was designated in any past year, so the statistic that would show a decade-long failure cannot be built.
• No post-obligation retention is published for the federal program. The service agreement runs three years and USDA publishes nothing at year four. The statement that about half of food-animal graduates leave over a career is a career-length observation about graduates generally, not an outcome measure for this program.
• No target exists. No document sets a number of designations, a fill rate, or a workforce share to reach. The August 2025 action plan lists five actions and no figure, and the appropriation is a dollar amount with no denominator.
• No cost of the counterfactual is estimated. Every source asserts that these shortages endanger animal health, public health and the food supply. None estimates what one unfilled designation costs, in detection delay, in producer loss, or in outbreak probability.

See the evidence

What was read directly and what came second hand differ sharply here, and the difference falls along a line that matters. The regulatory architecture is primary and was read end to end — 7 CFR part 3431 through the eCFR renderer, the bill text of S. 2215 and the Federal Register information-collection notice through govinfo, the 2022 Census of Agriculture table and the January 2026 cattle report as agency PDFs, and the four NIFA program pages. USDA speaks in its own words in the August 2025 action plan, distributed as an agency bulletin. Every headline count of designations, however, and every workforce and debt and pay figure, comes from one trade association, and the FY2026 funding notice was read as a reposting by a second association rather than from the agency. Two disagreements are left visible rather than resolved. The first is which designation set governs the FY2026 cycle — one source attributes the cycle to the FY2025 set of 243 areas across 46 states identified in the autumn of 2024, the reposted USDA notice says the 2026 situations are published, and the program page dates the posting to 2026-01-05, with no source reconciling the three. The second is the workforce category — 3,424 veterinarians in food-animal practice in 2024 and slightly more than 8,100 in food or mixed animal practice in 2023 come from the same publisher, count different things, and are never set against each other, so every sentence above names which category it uses. Three quantities in this document are arithmetic performed here and printed by no source, and each is marked as such where it appears — the ratio of about 1.75 producers per farm, the product of about 1,281,000 producers, and the subtraction of 456,536 operations between 1997 and 2022. Where a source could not be opened the row is kept with an empty URL and the reason, including the nomination form, whose full field list was read but only from an ephemeral staging mirror that is not recorded here as a durable link. This is a Path A output, so observation_refs is empty and provenance_mode: press-derived.

People affected

Estimated range 1,281,215–3,333,333 As of 2022 Census of Agriculture

Sensitivity This range is the national client base of food-animal veterinary practice. It is not a count of people inside a designated veterinary shortage area and it must not be read as one. The in-area count is not derivable and the missing input is specific. USDA publishes the 245 designations of 2026 as a map application rather than as a county roster, and 7 CFR 3431.7 places them on an agency web site rather than in the Federal Register, so no dated roster of designated counties exists for any year. Designations sit in 47 of 50 states, which makes the national base the correct universe to bound, but the share of that base inside a designated area cannot be computed from any published source and no proxy share is defensible, so none is offered. The width between the bounds is the difference between cattle and calves operations alone and all farm producers of every commodity. In the opposite direction the range still fails to count several groups entirely. The general stake in animal-disease surveillance and in detection lead time for a foreign animal disease is asserted by USDA and quantified nowhere. The animals are not counted, and national cattle inventory stood at 86.2 million head on 2026-01-01. Meat and poultry inspection and export certification capacity sits behind public-practice designations whose clients are regulatory functions rather than farms. Rural residents who are not producers but live in a county whose only large-animal practice closed fall outside every count used here. The 3424 veterinarians in food-animal clinical practice in 2024 are excluded on purpose, because they are the missing service rather than the affected population.

Regional breakdown USDA publishes veterinary shortage designations as a map application and, under 7 CFR 3431.7, on an agency web site rather than in the Federal Register. No county roster for the 245 designations of 2026 was reachable this round, and the per-state maximum nomination allocation table returned HTTP 403 on every route tried, so no state or regional split of the affected base can be confirmed.